Build QoE 2.0 for Your SaaS 18 Months Before You List

Your EBITDA Bridge Is No Longer Enough

Direct answer: Traditional quality of earnings review, normalizing EBITDA, flagging non-recurring items, adjusting owner compensation, is table stakes in 2026. PE buyers now request what one M&A analyst calls "QoE 2.0" before they'll even discuss an LOI: revenue broken out by customer cohort, churn that's being masked by aggregate growth, margin analyzed by contract type, net revenue retention by customer vintage, and a full map of which parts of your revenue depend on AI tools you don't control. A 2026 analysis of PitchBook's Q1 middle-market data lays out exactly what this new diligence standard requires, and founders who build it 18 months before they list control the narrative. Founders who wait get it built for them, on the buyer's timeline, with the buyer's conclusions.

In the Navy, we had a term: "qual card." Every system on the submarine had a qualification card, a documented checklist proving you understood that system inside out. QoE 2.0 is your business's qual card. If you can't walk a buyer through every revenue cohort, every churn pattern, every margin driver, you haven't qualified your own business.

What PitchBook's Q1 2026 Data Actually Signals

The PitchBook Q1 2026 middle-market report shows deal value at $103.8 billion with a median deal size of $193.1 million, and sponsor-to-sponsor transactions accounting for 69.5% of exits. That last number matters more than it looks. The IPO channel has effectively closed for mid-market software companies, and corporate acquirers hold only about 30.5% share. That means most SaaS exits now run through professional buyers who do this full-time, who have seen every seller trick, and who are no longer satisfied with the diligence package that used to close deals three years ago.

The report's core argument is that quality of earnings has quietly split into two tiers. Traditional QoE answers "is the reported EBITDA real." QoE 2.0 answers a harder question: "is the growth durable, and can you prove it at the cohort level instead of the aggregate level." Sponsor buyers doing 69.5% of deals are running this second tier of analysis whether or not the seller volunteers the data, because their own investment committees now require it before capital gets deployed.

The Five Things QoE 2.0 Actually Checks

Revenue by customer cohort. Not total ARR. ARR broken into cohorts by signup date, showing how each cohort's revenue trends over its lifetime. A single blended growth number can hide a business where every cohort after month eighteen starts shrinking, and the aggregate number only looks healthy because new cohorts keep entering at the top.

Churn masked by aggregate growth. The same principle applied specifically to logo and revenue churn. CT Acquisitions' 2026 SaaS valuation data is explicit that net revenue retention below 100% drops the multiple by 1x to 3x ARR regardless of the headline growth rate, and buyers now check whether your reported NRR is a blended average hiding a weak core segment behind strong expansion revenue from a handful of large accounts. SaaS Capital's 2026 survey of more than 1,000 private B2B SaaS companies puts median gross revenue retention at 91% for bootstrapped companies in the $3 million to $20 million ARR range, with a median NRR of 103%. If your blended NRR sits at 103% but your GRR is well below 91%, your expansion revenue is quietly covering for churn a buyer's team will find the moment they decompose the number.

Margin by contract type. Enterprise contracts with heavy implementation services carry different margin profiles than self-serve subscriptions. A buyer wants to see both separately, because a business that looks like 80% gross margin software can actually be a lower-margin services business wearing a SaaS label if enough revenue comes bundled with delivery labor.

NRR by vintage. Not just current NRR. NRR trended across cohorts by signup year. A business whose 2022 cohort retains at 115% but whose 2025 cohort retains at 92% is telling you something about product-market fit or competitive pressure that a single blended NRR figure will never surface.

AI-dependency mapping. Which parts of your product, your support function, and your go-to-market motion depend on a third-party AI model or API you don't control. Acquinox Advisors' 2026 software M&A research flags AI-driven commoditization as a live threat to undifferentiated horizontal software specifically, meaning buyers now ask whether your product's defensibility survives if a foundation model vendor ships a competing feature natively. Vertical specialization and genuine AI integration earn a premium. A thin wrapper around someone else's model does not.

Per-Acquisition Baseline-and-Uplift Attribution

The same report introduces a requirement several sponsor buyers are now writing into diligence checklists: per-acquisition baseline-and-uplift attribution. In plain terms, for every customer acquisition channel, show what performance would have looked like without any AI-driven optimization, versus what actually happened with it. If you can't isolate the uplift AI specifically contributed to a channel's performance, from the baseline that channel would have delivered anyway, you can't prove your AI investment is doing anything measurable. Buyers who ask this question are testing whether your growth story is real optimization or a coincidence you're calling optimization.

This is uncomfortable for a lot of founders because most teams never built the tracking to answer it. Build it now, eighteen months out, not during diligence when you're trying to reconstruct historical attribution from memory and half-complete dashboards.

Why Eighteen Months, Not Six

Cohort-level data takes time to accumulate meaningfully. If you start building the tracking infrastructure six months before you list, you'll have six months of clean cohort data and a pile of historical data you have to reconstruct or, worse, admit you don't have. Eighteen months gives you enough forward-looking clean data that a buyer can see trends rather than a single snapshot, and enough runway to fix whatever the early data reveals. If your month-eighteen cohort retention is weak, eighteen months is enough time to test a fix and show the buyer the fix worked, which is worth far more than showing them the problem with no response to it.

Rule of 40 Is Not the Safety Net It Used to Be

Acquinox Advisors' research notes that hitting the Rule of 40, combined growth rate and profit margin above 40%, used to be considered a safety net against sale pressure. That dynamic is shifting because AI has changed the underlying economics of software delivery. A company hitting Rule of 40 through legacy efficiency gains, without genuine AI-driven differentiation, is increasingly viewed as vulnerable to commoditization rather than as a safe, stable asset. The metric alone no longer tells a sophisticated buyer what they need to know. They want the cohort data and the AI-dependency map underneath it.

Buyers Are Rebuilding Your Cohort Curves Whether You Help or Not

CT Acquisitions' 2026 SaaS buyer's playbook confirms sophisticated buyers no longer accept a seller's aggregate retention number at face value. Experienced buyers now pull customer-level data going back 24 to 36 months and rebuild cohort curves from scratch, segmented by contract value band, industry, acquisition channel, and product tier. A flat retention curve landing at 95% is treated as healthier than a steeper curve that also lands at 92% blended, because the shape tells a buyer whether the business is stabilizing or still eroding. If you hand over aggregate numbers and let the buyer's team build the cohort curve themselves, you've ceded the entire narrative to whatever conclusion their model reaches first, and that conclusion rarely favors the seller.

The same rebuild happens to your net dollar retention figure. A 115% NDR built on 90% gross retention plus 25% expansion revenue is a fundamentally different, more fragile business than a 115% NDR built on 95% gross retention plus 20% expansion, even though the headline number is identical. Buyers decompose the blend because the blend hides which piece is doing the work. Do this decomposition yourself before they do it to you, and you control which story gets told first.

Building the QoE 2.0 Package Yourself

Start with your billing and usage data warehouse, not your accounting system. Traditional QoE lives in the general ledger. QoE 2.0 lives in your product analytics and CRM, cross-referenced against billing. Pull every customer's signup date, plan tier, expansion and contraction history, and support ticket volume, then group by cohort and by contract type. If your data infrastructure can't produce this cut cleanly today, that's the first gap to close, well before you engage a banker or start buyer conversations.

Bring in a fractional CFO or M&A advisor who has specifically built QoE 2.0 packages before, not a generalist accountant. CT Acquisitions' guidance on SaaS sale preparation names customer concentration, gross margin, CAC payback, and Rule of 40 compliance as the specific metrics buyers focus on alongside NRR and GRR. Your package needs to answer each of these with cohort-level granularity, not a single blended figure per metric.

Doctrine Connection: Verification Beats Optimism

Every founder believes their growth story. That belief is worth nothing to a buyer without evidence that survives scrutiny at the cohort level. Optimism is what gets a pitch deck built. Verification is what gets a deal closed at the multiple you were hoping for instead of the multiple the buyer's diligence team decides to hand you after finding the gaps themselves. The qual card doctrine applies here exactly as it applied on the boat: you don't get to claim you understand a system. You demonstrate it, checklist by checklist, until there's no ambiguity left for someone else to exploit. Build the proof before someone else demands it, on their schedule, framed as a problem instead of a strength.

FAQ

Q: Is QoE 2.0 only relevant for venture-backed SaaS companies, or does it apply to bootstrapped businesses too?
It applies to both. Bootstrapped SaaS companies selling to PE buyers face the same sponsor-driven diligence standard described in the PitchBook data, since 69.5% of deals now run sponsor-to-sponsor or through PE platforms regardless of how the target company was originally funded. The size of the check doesn't change the diligence standard nearly as much as founders expect.

Q: What's the single biggest gap founders have when they first try to build this?
Cohort-level revenue tracking. Most SaaS founders can produce a clean blended MRR or ARR chart in minutes. Very few can immediately produce revenue and retention broken out cleanly by signup cohort, because most billing systems weren't built with that reporting in mind from day one. Fixing this is usually the first and most time-consuming step.

Q: How does AI-dependency mapping differ from a general technology risk assessment?
It's narrower and more specific. A general tech risk assessment covers infrastructure, security, and technical debt broadly. AI-dependency mapping asks one focused question: which parts of your product's value proposition depend on a third-party AI model or API you don't own, and what happens to your differentiation if that vendor ships a competing feature or changes pricing.

Q: Can I build QoE 2.0 internally, or do I need outside help?
You can start internally, especially the cohort data infrastructure, since your own team understands your billing system better than an outside advisor will on day one. Bring in outside help for the analysis and framing once the raw data exists, ideally someone who has built these packages for other SaaS exits and knows what a sponsor buyer's diligence team will actually ask for.

Q: What happens if I skip this and go to market with only traditional QoE?
You'll likely still get offers, but expect a longer diligence period, more re-trade risk after the LOI, and a real chance the buyer's own analysis finds gaps you didn't disclose, which weakens your negotiating position at exactly the moment you have the least room to give. Building QoE 2.0 yourself, in advance, converts that same information from a liability into a selling point.

Disclosure: Jeff Barnes, MBA has no personal position in any company, fund, or platform named in this article. demg.ai provides marketing education and systems for owner-operators, not investment advice.